RuntimeError: expected scalar type Long but found Int #1989
Comments
|
This should have been fixed (#1954) in the 0.21.3 release. Can you make sure you're using the latest version of Raster Vision? |
|
yes ,i run the pip list and check the Raster Vision, it is 0.21.3 |
|
That's strange. Please share your code. |
|
I run the train.ipynb,and change the train_image_uri 、train_label_uri、val_image_uri、val_label_uri as my local data path and get the error,and my torch is 2.1.0+cu121 |
|
I am not able to reproduce the error using version 0.21.3 and running the code in |
|
I ran into this issue as well and found that, in my dataset, adding an albumentations transform like |
Same issue without change the URLs. |
How to add transfom,can you share you code snipe?thanks |
|
See this discussion for an example |
|
I change the OS to ubuntu 22.04,and no issue happen,it works well. |
f:\raster-vision\docs\usage\tutorials\train.ipynb cell 41 line 1
----> 1 learner.train(epochs=3)
File d:\Python310\lib\site-packages\rastervision\pytorch_learner\learner.py:391, in Learner.train(self, epochs)
389 for epoch in range(start_epoch, end_epoch):
390 log.info(f'epoch: {epoch}')
--> 391 train_metrics = self.train_epoch(
392 optimizer=self.opt, step_scheduler=self.step_scheduler)
393 if self.epoch_scheduler:
394 self.epoch_scheduler.step()
File d:\Python310\lib\site-packages\rastervision\pytorch_learner\learner.py:415, in Learner.train_epoch(self, optimizer, step_scheduler)
413 batch = (x, y)
414 optimizer.zero_grad()
--> 415 output = self.train_step(batch, batch_ind)
416 output['train_loss'].backward()
417 optimizer.step()
File d:\Python310\lib\site-packages\rastervision\pytorch_learner\semantic_segmentation_learner.py:30, in SemanticSegmentationLearner.train_step(self, batch, batch_ind)
28 x, y = batch
29 out = self.post_forward(self.model(x))
---> 30 return {'train_loss': self.loss(out, y).long()}
...
3051 if size_average is not None or reduce is not None:
3052 reduction = _Reduction.legacy_get_string(size_average, reduce)
-> 3053 return torch._C._nn.cross_entropy_loss(input, target, weight, _Reduction.get_enum(reduction), ignore_index, label_smoothing)
RuntimeError: expected scalar type Long but found Int
The text was updated successfully, but these errors were encountered: